With the increasing popularity of “telecommuting” virtual private networks (VPNs) are increasingly being used by organizations to connect remotely located (e.g. staff) computers to the organization's local area network (LAN). VPNs are advantageous in that they do not require substantial hardware to implement as the “virtual network” is created across a pubic network, whilst also providing a relatively high level of security.
However, when creating Layer 2 VPNs of this type, the remote computer and the in-house LAN must share a common IP address space, as a single subnet. If the remote computer is a part of another network e.g. it is part a separate LAN, sharing a common IP address space is difficult and usually requires re-addressing the remote computer and those of the organisation into a common private or public address space. The inventors of the present invention have identified that this situation is not optimal and that in certain circumstances it would be preferable to allow the organization and the remote device (or its associated network) to maintain their original IP address space.